Disclosing the bioactive metabolites involved in the in vitro anthelmintic effects of salt-tolerant plants through a combined approach using PVPP and HPLC-ESI-MSn

Strategies to reduce dependence on synthetic drugs for the treatment of gastrointestinal nematodes (GIN) infections in ruminants include the search for novel anthelmintic scaffolds on plants, yet salt-tolerant plants remain overlooked. This study aims to evaluate the in vitro anthelmintic properties of selected salt-tolerant plants against GIN, and identify the potential bioactive secondary metabolites involved. For that purpose, 80% acetone/water extracts were prepared from dried biomass of aerial organs of nine salt-tolerant plant species and tested against Haemonchus contortus and Trichostrongylus colubriformis by the Larval Exsheathment Inhibition Assay (LEIA) and Egg Hatching Inhibition Assay (EHIA). Pistacia lentiscus, Limoniatrum monopetalum, Cladium mariscus and Helychrisum italicum picardi were the most active in both GIN and life stages. To investigate the role of polyphenols in the anthelmintic activity, four selected extracts were treated with polyvinylpolypyrrolidone (PVPP), and non-treated and treated samples were further characterized by high-performance liquid chromatography with electrospray ionization mass spectrometric detection (HPLC-ESI-MSn). While polyphenols seem responsible for the EHIA properties, they are partially accountable to LEIA results. Several phenolics involved in the anthelmintic effects were identified and discussed. In sum, these species are rich sources of anthelmintic compounds and, therefore, are of major interest for nutraceutical and/or phytotherapeutic applications against GIN in ruminants.

www.nature.com/scientificreports/ (10 mg mL −1 ) were mixed with 200 µL of a methanol solution of DMACA (1%, w/v), and 100 µL of hydrochloric acid (37%, v/v). After a 15 min incubation period, absorbance was measured at 640 nm in a multiplate spectrophotometer reader. A calibration curve was prepared using catechin as a standard. The concentration of CT was expressed as catechin equivalents (mg CE g extract −1 , DW).

Chemical profiling by high-performance liquid chromatography with electrospray ionization mass spectrometric detection (HPLC-ESI-MS n )
. HPLC-ESI-MS n analyses were performed with an Agilent Series 1100 HPLC system with a G1315B diode array detector (Agilent Technologies), and an ion trap mass spectrometer (Esquire 6000, Bruker Daltonics) with an electrospray interface. Separation was performed in a Luna Omega Polar C 18 analytical column (150 × 3.0 mm; 5 µm particle size) with a Polar C 18 Security Guard cartridge (4 × 3.0 mm), both purchased from Phenomenex. Detailed chromatographic conditions are available in Supplementary Material files. Compounds' identification was performed by mass spectrometry data. Compounds' quantitation was carried out by UV using analytical standards of neochlorogenic acid (320 nm), chlorogenic acid (320 nm), protocatechuic acid (280 nm), catechin (280 nm), sinapic acid (320 nm), ferulic acid (320 nm), quercetin (350 nm), apigenin (350 nm), and kaempferol (350 nm). Detection limits (3σ criterion) ranged between 0.06 and 0.15 mg L −1 . Calibration graphs were constructed in the 0.5-100 mg L −1 range. Peak areas at the corresponding wavelengths were plotted versus analyte concentration. Each analytical standard was used to quantify the corresponding compounds or compounds of the same chemical family for which the exact analytical standards were not available. Repeatability (n = 10) and intermediate precision (n = 9, three consecutive days) were lower than 4 and 8%, respectively. The robustness of the chromatographic method was evaluated by recording analyte signals at ± 2 nm of the optimum wavelength and by slightly varying the percentage of the mobile phase (2% changes), observing variations lower than 5% for all the analytes concerning the optimum conditions.
In vitro anthelmintic assays. Haemonchus contortus and Trichostrongylus colubriformis parasites. Thirdstage larvae (L3) and eggs were obtained from faeces of monospecifically infected caprine and ovine donors, with susceptible strains of H. contortus and T. colubriformis. L3 larvae had been maintained in culture flasks for 1 month, at 4 °C, before use in the Larval Exsheathment Inhibition Assay (LEIA), while eggs were collected on the day of the Egg Hatching Inhibition Assay (EHIA), and used up to 2 h after collection 24 .
LEIA. LEIA was performed as previously described by Bahuaud and colleagues 25 . The extracts were diluted in phosphate-buffered saline (PBS; 0.1 M phosphate, 0.05 M NaCl, pH 7.2), and incubated with L3 larvae (approx. 800 larvae per mL) at 23 °C for 3 h. Afterwards, larvae were washed and centrifuged with PBS 3 times, and the pellet resuspended in 200 µL of PBS. To initiate the LEIA, 40 µL of the test solution was used to count the proportion of ensheathed/exsheathed larvae at 0 min. The remaining larvae (160 µL) were then subjected to an artificially induced exsheathment by exposure to a solution of Milton (2% w/v sodium hypochlorite, 16.5% w/v sodium chloride) diluted in PBS. Milton optimal concentration was determined for each batch before use in order to ensure a gradual exhsheathment process, reaching 100% exsheathment in 60 min. After 20, 40, and 60 min exposure, the number of ensheathed and exsheathed larvae were counted under a microscope (400×). Four replicates were performed for each extract concentration, and the negative control, PBS, was run in parallel. Percentage of larvae exsheathment (LE) for each replicate was calculated according to the following formula: %LE = [(number of exsheathed larvae)/(number of exsheathed + ensheathed larvae) × 100].
EHIA. Faeces material was filtrated using a gaze hydrophyle compress for 2 times, transferred to a 25 µm sieve, and washed with distilled water. The residue was centrifuged three times using a saline saturated solution (d = 1.2) to concentrate the eggs, and the pellets were recovered in PBS for use in the experiments. Afterwards, the eggs were quantified, plated in 48-well sterile plates (100 eggs per well), and exposed to the extracts at concentrations ranging from 5000 to 78 µg mL −1 in PBS. Plates were incubated at 27 °C for 48 h in the dark, and the number of larvae and eggs, in each well, was registered after microscopic counting. Six replicates were performed for each extract concentration, and the negative PBS control was run in parallel. The percentage of egg hatching Polyvinylpolypyrrolidone (PVPP) treatment. PVPP is a polyphenol inhibitor that binds to tannins and flavonoids, removing these metabolites from the solution 26 . To ascertain the role of the polyphenols in the anthelmintic activity of the extracts, PVPP was added at a ratio of 50:1 to the active ones, respectively for eggs and larvae assays, in PBS, and incubated overnight at 4 °C. The maximum concentration tested for LEIA was 1200 µg mL −1 ; for EHIA it was 2500 µg mL −1 . Thereafter, the samples were centrifuged for 10 min at 4500 rpm, and the supernatant depleted in polyphenols was tested in LEIA and EHIA assays. The extracts exposed or not to PVPP plus a negative control (PBS) were run in parallel.

Statistical analyses.
At least four replicates per concentration were included in all experiments. The results on phenolic content are expressed as mean ± standard error of the mean (SEM). Anthelmintic data are expressed as the concentrations inhibiting 50% of larval exsheathment or egg hatching (IC 50 values, µg mL −1 ), and 95% confidence intervals (CI), obtained by Probit analysis. SPSS Statistics v. 26.0 software was used to assess significant differences among IC 50 values, through relative median potency estimates, and among phenolic data, by one-way analysis of variance (ANOVA) followed by the post-hoc Tukey HSD test. Spearman correlations were calculated between the total flavonoids, total phenols, and the IC 50 values for LEIA on the 2 nematode species.

Results and discussion
Total phenolics, total flavonoids and condensed tannins contents. The phenolic contents of the extracts are presented in Table 2. The phenolic content of the extracts is presented in Table 2. The total phenolic content of all species ranged between 14.2 and 226.3 mg GAE eq. g −1 DW extract while the total flavonoid content ranged between 13.3 and 45.4 mg QE g −1 DW. Lopes and colleagues (2016) reported higher TPC values for C. mariscus (254 mg GAE g −1 DW), C. soldanela (144 mg GAE g −1 DW), I. crithmoides (141 mg GAE g −1 DW), L. monopetalum (248 mg GAE g −1 DW) 80% acetone water extracts, except for P. lentiscus (130 mg g −1 DW), but lower flavonoid contents in comparison to our work (1.26-13.8 mg rutin g −1 DW) 15 . In another work, H. italicum picardi infusions and decoctions of aerial organs have been previously described as rich sources of flavonoids (91.8-119 mg rutin 200 mL −1 ) 27 . Moreover, a lower combined TPC value was detected in P. coronopus leaves and flowers extracts of different polarities (72.1 mg GAE g −1 DW) but increased TFC levels (282.8 mg rutin g −1 DW) 28 . In this study, total condensed tannins were detected only in three species, in the following concentration order: P. lentiscus > L. monopetalum > C. mariscus. In agreement, tannins were formerly detected in the same three formerly mentioned species, although at lower concentrations (6.63-38.7 mg CE g −1 DW, extract) 15 . It is worth to mention that dissimilarities between our results and those of other authors may be the reflection of different extraction methodologies and standards employed as well as environmental and plant-related factors.
In vitro anthelmintic properties. Table 3 summarizes the results of the in vitro activity of salt-tolerant plant extracts against H. contortus L3 larvae and eggs and T. colubriformis L3 larvae and eggs obtained in LEIA and EHIA assays. Lentisk (P. lentiscus) exhibited the highest activity on LEIA (IC 50 = 27.8-29.7 µg mL −1 ) and egg hatching processes (IC 50 = 197.7 and 223.9 µg mL −1 ), without significant differences between GIN species. Lentisk is an evergreen shrub with high polyphenol content and previous results have shown both in vitro and in vivo anthelmintic properties [29][30][31][32] . In previous studies, P. lentiscus extracts (acetone, ethanol and/or water) exhibited less than 20% larvae exsheathment and migration at 1200 µg mL −129,31 . Nevertheless, the results for the in vitro egg hatching assay are herein, to the best of our knowledge, described for the first time.  (Table 3). Limoniastrum monopetalum is a highly salttolerant shrub, widely distributed in the Mediterranean area, and was as effective as P. lentiscus in LEIA (p < 0.05), with IC 50 values lower than 50 µg mL −1 (no significant difference between the two tested parasites; p > 0.05). In EHIA, L. monopetalum was also the most active species, besides P. lentiscus, with similar activity towards both parasites (IC 50 = 1999.9 and 2102.5 µg mL −1 , respectively). Cladium mariscus, or sawgrass, is an evergreen grasslike plant occurring in coastal saltmarshes in the Mediterranean region. C. mariscus extract inhibited L3 larvae exsheathment (IC 50 = 77.8-88.9 µg mL −1 ), without significant differences between both parasite species (p > 0.05). In contrast, in the EHIA, C. mariscus was more effective towards H. contortus (IC 50 = 1496.6 µg mL −1 ) than T. colubriformis (IC 50 = 2575.5 µg mL −1 ; p < 0.05). Helichrysum italicum subsp. picardi (everlasting) is an aromatic salt tolerant plant commonly found in sandy soils, such as sand dunes, along the Southern European coast. Everlasting extract exhibited IC 50 values ranging between 92.8-132.5 µg mL −1 on LEIA, and 2947.7-3707.5 µg mL −1 on EHIA. Interestingly, H. contortus larvae and eggs were more susceptible to the H. italicum picardi extract than those of T. colubriformis (p < 0.05).
It is well recognized that the anthelmintic activity is affected by the class, structure and concentration of secondary metabolites 7 . Moreover, these metabolites have different effects, depending on the target parasite species and life development stages 7 . A higher susceptibility of H. contortus in comparison to T. colubriformis, as observed for C. mariscus and H. italicum picardi extracts, has been previously documented for other bioactive plants, such as sainfoin, and individual chemical structures, depending on the ratios of prodelphinidins/procyanidins 10,33,34 . The authors suggest that such differences can reflect dissimilarities on the composition of specific parasite sheath proteins, that interact differently with the chemical groups 33,34 . The same conclusion can be driven for differences among parasite stages, as the eggshell and larvae coat differ in their structural components, which has also been recorded with conventional anthelmintic drugs 7,35 . This may explain the results obtained for P. coronopus, which was more active against larvae exsheathment (IC 50 = 94.0 and 212.4 µg mL −1 ), and inactive towards eggs, of both parasite species, at the maximum concentration tested. Overall, IC 50 results obtained in LEIA are frequently reported as lower than EHA, suggesting that infective L3 larvae are more susceptible than eggs 36,37 .
Calystegia soldanella, C. maritima and M. marina co-occur in sand dunes along the Algarve coastline while I. crithmoides can be found in highly saline environments, such as saltmarshes. These four species were mildly to poorly active on both assays (Table 3). Interestingly, while I. crithmoides was mostly ineffective in this study, its related species, I. viscosa 70% ethanolic extract exhibit anthelmintic properties against the larvae exsheathment of a mixture of Teladorsagia circumcincta and T. colubriformis parasites 32 , suggesting significant chemical diversity among the genus.
Overall, the nine plant extracts had comparable effects between the two GIN species (Spearmen correlation; R 2 = 0.96; p < 0.01). In addition, a negative correlation between the total phenolic content and the anthelmintic activity was noted, particularly with H. contortus parasites (Spearmen correlation; R 2 = 0.783; p < 0.05), suggesting that these metabolites may be involved in the anti-parasitic nematode's effects.
Role of polyphenols in the anthelmintic activity: PVPP as a polyphenol binding agent. In order to ascertain the role of polyphenols in the anthelmintic properties, the four plant extracts presenting results for both LEIA and EHIA were selected for further studies using PVPP. PVPP is a polyphenol inhibitor, as it binds to tannins and flavonoids, removing these metabolites from the solution 26 . Thus, if after PVPP exposure a loss of the anthelmintic activity is observed, it can be assumed that polyphenols are most probably responsible for the activity once they were formerly removed.
The effects of the addition of PVPP to extracts on EHIA and LEIA are illustrated in Figs. 2 and 3, respectively. The application of all the extracts with PVPP largely restored the egg hatching process (Fig. 2) to control values, suggesting that polyphenols are most probably involved in the inhibition of this life stage development. www.nature.com/scientificreports/ Vargas-Magaña and colleagues (2014), while exploring the role of polyphenols on the anthelmintic effects of several extracts of tannin-containing tropical plants on EHIA, concluded that the main mechanism of action was by impairing larvae eclosion from the eggs 38 . Likewise, we noted a high number of larvae trapped inside the eggs after the application of these active extracts (data not shown).
In contrast to EHIA, results with PVPP varied on LEIA (Fig. 3): the application of the L. monopetalum extract, resulted in 60-70% of larvae exsheathment of both parasite species after PVPP addition for 60 min, in contrast to 0% in the non-treated sample; the extract from H. italicum picardi pre-incubated with PVPP remained mostly completely active. Subtle changes were observed for C. mariscus (approx. 20-40% of larvae exsheathment after 60 min of treatment) for both parasite species, while P. lentiscus had only around 20% of larvae exsheathment at 60 min, after PVPP treatment. These results suggest that other bioactive metabolites, alone or in synergy, can be present in all extracts tested, especially for H. italicum picardi, P. lentiscus, and C. mariscus. In agreement with our results, other authors already reported that P. lentiscus extracts remain active on GIN larvae migration after exposure to PVPP 29 .
The remaining activity on LEIA for the majority of the extracts tested should be carefully analyzed, and two scientific questions arise. First, was the ratio of PVPP used insufficient to cope with the high phenolic content of the extracts? Despite being commonly used, Manoloraki et al. (2010) questioned this hypothesis when testing P. www.nature.com/scientificreports/ www.nature.com/scientificreports/ lentiscus for larvae migration after PVPP addition, since this species has a high polyphenol content, comparable to our results 29 . On the other hand, are other bioactive metabolites present in the extracts that are also effective in inhibiting larvae exsheathment? For instance, different authors suggest that terpenes may be responsible for the remaining in vitro and in vivo anthelmintic properties of P. lentiscus after the addition of PVPP or polyethylene glycol (PEG), a similar inhibitor of polyphenols 29,30 . Additionally, Botura and colleagues (2013) described that the flavonoid fraction of Agave sisalana Perrine (sisal) had higher activity on egg hatching, while the saponin fraction had mostly larvicidal effects 39 . In an attempt to address these scientific questions, and elucidate the possible metabolites involved, we have conducted an HPLC-ESI-MS n comparative analysis on the active samples, before and after PVPP treatment. Obtained chromatograms are represented in Fig. 4 while the chemical profile of each species is depicted in Tables 4, 5, 6 and 7. The characterization of the compounds is detailed in Supplementary Material files.

HPLC-ESI-MS
The main constituents of P. lentiscus extract were flavonoid glycosides (mainly from myricetin and quercetin; approx. 53 mg g −1 DW), and galloylquinic acid and di-O-galloylquinic acid isomers (60 mg g −1 DW; Table 4; Suppl. files, Table I). In agreement to our findings, Romani et al. (2002) detected a high concentration of galloyl derivatives (5.3% DW), and a substantial amount of myricetin and quercetin glycosides (1.5% DW), extracted from a 70% ethanol solution of leaves 40 . Hydrolysable tannins are a group of gallic acid esters associated with polyols (e.g., glucose, glucitol, quinic acid), and the etherification or oxidation of the galloyl groups leads to the formation of complex structures (gallotannins and ellagitannins) 41 . Plant extracts containing hydrolysable tannins with gallic acid units were more effective as anthelmintics than those containing condensed tannins 42 . Nevertheless, the oligomerization and molecular weight of tannins may affect the anthelmintic activity, as is the case, for example, of elagitannins and condensed tannins 34,43 . Other metabolites present in lower concentrations in P. lentiscus extract with reported anthelmintic effects include flavan-3-ols and its galloyl derivatives, namely epigallocatechin (6.4 mg g −1 DW), gallocatechin gallate (6.8 mg g −1 DW) and catechin (5.0 mg g −1 DW). Molan et al. (2003) found that the presence of the galloyl group on flavan-3-ols was crucial for the activity on T. colubrifomis egg hatching (20% vs. 100% inhibition at 1 mg mL −1 ), and also more effective on immobilizing infective larvae (100% inhibition at 100-150 µg mL −1 ) 44 .
In P. lentiscus PVPP-treated samples, the concentration of flavonoid glycosides (0.17 mg g −1 DW) and galloylquinic acid (2.2 mg g −1 DW) drastically dropped (Suppl. files, Table I), which may justify the restoration of the egg hatching. On the other hand, the presence of these compounds in lower concentrations may explain the remaining activity on larvae. Nevertheless, compounds 2, 58, and 78 remained in this sample, and may also account for the activity.
Caffeoylquinic and dicaffeoylquinic acids were the most abundant compounds in H. italicum picardi extract (150 mg g −1 DW), followed by quercetin-O-glucosides (approx. 31 mg g −1 DW; Table 5; Suppl. files, Table II). These findings were expected, since previous works identified high contents of these metabolites in aerial organs of the same species 27,45 . Borges and colleagues (2019) found a significant correlation between the phenylpropanoid content (particularly chlorogenic acid, 1,3-dicaffeoylquinic, and 3,5-dicaffeoylquinic acids), and the ovicidal activity of 17 plant extracts from Pantanal wetlands against Haemonchus placei 46 . Additionally, chlorogenic acid exhibited an IC 50 value of 92.4 μg mL −1 against L3 larvae exsheathment of H. contortus, and was also effective on preventing larvae hatching from eggs (IC 50 = 520.8 μg mL −1 ) 47 . These results point out the potential of caffeoylquinic and dicaffeoylquinic acids to be the active metabolites of H. italicum picardi extracts. However, some O-glycosides are also present that may contribute to the detected activity. For example, Barrau and colleagues (2005) tested the activity of 3 flavonol glycosides (quercetin-3-O-rutinoside or rutin, kaempferol-3-rutinoside or nicotiflorin, and isorhamnetin-3-rutinoside or narcissin), and all reduced the migration of H. contortus L3 larvae in 25-35% when applied at 1200 µg mL −148 .
In H. italicum picardi PVPP-treated sample, although in lower concentrations, caffeoylquinic and dicaffeoylquinic acids remained in solution (8.3 mg g −1 DW), from which chlorogenic acid was the main compound (6.3 mg g −1 DW; Suppl. files, Table II). The high activity observed for the extract from H. italicum picardi treated with PVPP on larvae exsheathment is most likely due to the high content of chlorogenic acid remaining in the sample 47 . Still, other caffeoylquinic and dicaffeoylquinic acids are present (2 mg g −1 DW) that might also add to its effects. On the other hand, in EHIA the lower amount of these compounds in the PVPP-treated sample may have not be sufficient to inhibit egg hatching, since this process was completely restored. In fact, Borges and colleagues (2019) suggest that the concentration of monomeric and dimeric chlorogenic acid derivatives that enter in contact with eggs seems to be determinant for the activity, as observed for Melanthera latifolia ethanolic extract that had low concentrations of these compounds. and was considered inactive (up to 80% egg hatching at 50 mg mL −1 ) 46 .
Cladium mariscus acetone water extracts were previously reported as a rich source of polyphenols, particularly tannins by spectrophotometric methods, and chlorogenic, ferulic, and syringic acids were detected in higher amounts, through HPLC-DAD analysis 15,49 . In agreement, in this study, C. mariscus extract was mainly composed of flavan-3-ols (epigallocatechin, catechin), proanthocyanidins (5.1 mg g −1 DW), luteolin, C-glycosyl luteolin, a kaempferol glucoside, and an apigenin flavone (9.5 mg g −1 DW; Table 6; Suppl. files, Table III). Flavan-3-ols and proanthocyanidins have well recognized anthelmintic effects 44,50 , and therefore, they are most likely involved in the activity of C. mariscus extract. Also, the activity of the flavonoid luteolin on H. contortus larvae exsheathment has been previously established (IC 50 = 17.1 and < 71.5 µM) 51 . Interestingly, Klongsiriwet and colleagues (2015) found that luteolin, even at low concentrations (30 μM), display synergistic effects with procyanidins, leading to a fivefold lower IC 50 of the mixture in comparison to the procyanidin fraction alone (75.9 vs. 356 μg mL −1 ) 51 . www.nature.com/scientificreports/ Having this in mind, the combination of proanthocyanidins and luteolin in C. mariscus extract could act synergistically in the inhibition of the egg hatching. Nevertheless, the activity on LEIA was only partially restored after PVPP addition (approx. 20-40% larvae exsheathment), i.e., the remaining metabolites are still exhibiting anthelmintic properties. In PVPP-treated samples, mainly C-glycosyl flavones (1.07 mg g −1 DW) and, to a less extent chlorogenic acid, remained in solution while the catechin derivatives and luteolin were removed (Table 6; Suppl. files, Table III). As previously addressed, chlorogenic acid exhibits significant anthelmintic activity in vitro against H. contortus larvae exsheathment and egg hatching 47 . Despite the activity described for luteolin, the www.nature.com/scientificreports/ www.nature.com/scientificreports/ investigation of the anthelmintic properties of its glycosides is lacking. In general, C-glycosyl flavones exhibit antioxidant and anti-inflammatory properties 52 , and two flavone-C-glycosides namely isoschaftoside and schaftoside shown strong toxicity (LC 50 = 114.66 μg mL −1 and 323.09 μg mL −1 ) against the plant-parasitic nematode Meloidogyne incognita 53 . Moreover, it is worth noticing that compounds 20 and 39 are still unidentified, although present in PVPP-treated samples. Previous works identified several phenolic compounds in L. monopetalum extracts including gallic, vanillic, ferulic, syringic, p-hydroxybenzoic, protocatechuic, chlorogenic, and trans-cinnamic acids, and also quercetin, apigenin, amentoflavone, flavones, methyl gallate, and myricetin 54,55 . In the current work, the main metabolites identified in L. monopetalum extract were epigallocatechin, phenolic acids and derivatives, isorhamnetin sulfate, Table 5. Characterization of the compounds present in the extract of Helichrysum italicum picardi aerial organs. Column "PVPP" indicate if the compound was also present in the corresponding H. italicum picardi treated PVPP sample. www.nature.com/scientificreports/ pinoresinol, methylated flavonoids sulfate and two oxylipins (Table 7). However, some of the major compounds, namely the methylated flavonoids sulfate 67, 71, 76, and 80 were not identified, as well as the minor metabolites 33, 46, and 55. The production of sulfated metabolites by plants is pointed out as an evolutionary trait to thrive in aquatic saline habitats, and part of the plant heavy metal detoxification mechanism 56,57 . Indeed, L. monopetalum is a halophytic and metal accumulator shrub that thrives in saltmarshes under harsh biotic and abiotic stresses (e.g., tidal fluctuations, salinity, heavy metal soils, sunlight exposure, UV radiation). Sulfated phenolics were previously identified in other halophyte species, such as Limonium caspium (Willd.) Gams 58 and Halimione portucaloides (L.) Aellen 59 . The pharmacological interest in sulphated flavonoids increased in the last decades, mainly driven by its hydrophobic nature, and many reported biological activities, like anti-coagulant, anti-viral, antioxidant, anti-inflammatory, antimicrobial 60 . Besides epigallocatechin (9.46 mg g −1 DW), the concentration of isorhamnetin sulfate (65) was high in L. monopetalum (6.4 mg g −1 DW) as well as phenolic acids and its derivatives (10.3 mg mL −1 DW; 7, 16, 24, 32). Delgado-Nuñez and colleagues (2020) attributed the main anthelmintic effects of Prosopis laevigata Willd. M. Johnston to isorhamnetin, which caused 100% of mortality on H. contortus eggs at the lowest concentration tested (700 µg mL −1 ), being also effective towards larvae (IC 50 = 2.07 mg mL −1 ) 61 . The glycoside Table 6. Characterization of the compounds present in the extract of Cladium mariscus aerial organs. Column "PVPP" indicate if the compound was also present in the corresponding C. mariscus treated PVPP sample.  62 . Nevertheless, one should keep in mind that structural modifications, such as glycosylation, methylation, and sulfation, may affect the bioactivity observed. For example, the substitution by a sugar unit in the quercetin structure showed a twofold increase in the larvicidal activity of rutin 62 . Still, studies concerning the anthelmintic effects of sulphated phenolics are missing. Since these metabolites are the main suspects as bioactive components of L. monopetalum extract, it would be interesting for further works to be conducted, not only confirming the anthelmintic effects of isolated compounds but also clarifying the role of sulfate in structure-activity relationship studies. After PVPP treatment, the activity of L. monopetalum extract on larvae exsheathment was restored by approximately 60-70% to the control values. Although the remaining compounds may have contributed to the overall activity, the major anthelmintic effects were annulated. As some main metabolites of L. monopetalum (67, 71, 76, 80) remain to be identified and quantified, further studies on this species are required to completely understand its bioactive compound (s) and related anthelmintic properties. www.nature.com/scientificreports/

Concluding remarks
Due to the constant diffusion of resistance to synthetic anthelmintics in worm populations, the search for plants with antiparasitic activities and their bioactive metabolites that can be used for integrated control approaches of GIN, has expanded over the last 20 years 63 . Extremophile plants, in particular salt-tolerant species, may represent an untapped reservoir of anthelmintic compounds for such purpose. To the best of our knowledge, this study explores for the first time the in vitro anthelmintic properties of eight salt-tolerant species, namely H. italicum subsp. picardi, I. crithmoides, C. soldanela, C. mariscus, M. marina, P. coronopus, L. monopetalum, and C. maritima, against two GIN species and life stages. Pistacia lentiscus, L. monopetalum, C. mariscus, and H. italicum subps. picardi were the most active against both parasite species and life stages (eggs and L3) targeted. The comparative HPLC-ESI-MS n analysis coupled with the use of PVPP unraveled that different bioactive metabolites may be involved in the anthelmintic properties: flavonoid glycosides and galloylquinic acid isomers in P. lentiscus; caffeoylquinic and dicaffeoylquinic acids and quercetin glycosides in H. italicum picardi; proanthocyanins, phenolic acids, and luteolin in C. mariscus; and sulphated and/or methylated flavonoids in L. monopetalum. Further work should be pursued to complete the identification of the main metabolites of L. monopetalum, since this species exhibited the most promising results after P. lentiscus. As recently comprehensively reviewed by Spiegler et al. 65 and Liu and colleagues 64 , polyphenols have been the most extensively studied compounds regarding their anthelmintic effects but the number of other individual phenolic compounds and their structural diversity investigated is still limited, particurlary towards these two GIN species. Therefore, future work should focus on fully elucidate the activity of the main potential bioactive metabolites identified in this work, either alone and/or in synergy, and provide information on structure-activity effects. Still, the results obtained in this study for L. monopetalum, C. mariscus, and H. italicum subsp. picardi warrant further investigations on the potential use of these species either as nutraceutical and/or phytotherapeutic options and/or as sources of anthelmintic compounds against GIN in ruminants. www.nature.com/scientificreports/